unitprot annotations in rdf tag

assemble.sh

 #!/usr/bin/env bash
 
+# ------------------------- PARAMETERS TO SET -------------------------
 # Load parameters either from parameters.sh, or from a file provided as the first parameter of the script
 if [[ $1 == "" ]]; then
     source parameters.sh
@@ -8,9 +9,6 @@ else
 fi
 # ------------------------- PARAMETERS TO SET -------------------------
 
-
-# ------------------------- PARAMETERS TO SET -------------------------
-
 ASSOCIATIONS_DIR=associations/
 ASSOCIATIONS_DATA_DIR=$ASSOCIATIONS_DIR/data/
 EXTEND_DIR=${ASSOCIATIONS_DIR}/extend_network/
@@ -157,9 +155,17 @@ check_exit_code
 
 log "Pathways assembled into ${map_out_path}"
 
+log "Implanting UniProt annotations..."
+map_out_path_uniprot=${map_out_path/.xml/_unp.xml}
+$PYTHON_BIN ${MAP_GENERATOR_DIR}/utils/implant_annotations.py -m ${map_out_path} -v ${minerva_variants_out_path} > ${map_out_path_uniprot}
+check_exit_code
+log "UniProt annotations implanted"
+
+
 log "Trimming long strings..."
-map_out_path_trimmed=${map_out_path/.xml/_trim.xml}
-$PYTHON_BIN ${MAP_GENERATOR_DIR}/utils/trim_long_strings.py -i ${map_out_path} > ${map_out_path_trimmed}
+map_out_path_trimmed=${map_out_path_uniprot/.xml/_trim.xml}
+$PYTHON_BIN ${MAP_GENERATOR_DIR}/utils/trim_long_strings.py -m ${map_out_path_uniprot} > ${map_out_path_trimmed}
+check_exit_code
 log "Long string trimmed"
 
 log "Combining the map with overlays"

map_generator/utils/implant_annotations.py

@@ -20,17 +20,33 @@ def implant(map_path: str, variants_path: str) -> str:
 
     for elem in list(root.iter(get_full_name("species", "", namespaces))):
     # for elem in root.iterfind(get_full_name("layout", "layout", namespaces)):
-        name = elem.attrib["name"]
-        if name in gene_uniprot:
-            for uniprot_id in gene_uniprot[name]:
-                idb = ET.Element('bqbiol:isDescribedBy')
+        elem_name = elem.attrib["name"]
+        elem_id = elem.attrib["id"]
+
+        if elem_name  in gene_uniprot:
+            rdfs = list(elem.iter(get_full_name("RDF", "rdf", namespaces)))
+
+            if len(rdfs) > 0:
+                rdf = rdfs[0]
+            else:
+                rdf = ET.SubElement(elem, 'rdf:RDF')
+
+            rdf_descs = list(rdf.iter(get_full_name("Description", "rdf", namespaces)))
+            if len(rdf_descs) > 0:
+                rdf_desc = rdf_descs[0]
+            else:
+                rdf_desc = ET.SubElement(rdf, 'rdf:about', {"#{}".format(elem_id)})
+
+
+            for uniprot_id in gene_uniprot[elem_name ]:
+                idb = ET.SubElement(rdf_desc, 'bqbiol:isDescribedBy')
                 b = ET.SubElement(idb, 'rdf:Bag')
                 urn_key = 'urn:miriam:uniprot' if '-' not in uniprot_id else 'urn:miriam:uniprot.isoform'
                 ET.SubElement(b, 'rdf:li', {'rdf:resource': "{}:{}".format(urn_key, uniprot_id)})
-                elem.append(idb)
 
 
-    return ET.tostring(root, encoding='utf8').decode('utf8')
+    # return ET.tostring(root)
+    return ET.tostring(root, encoding='unicode')
 
 if __name__ == '__main__':
 
@@ -42,7 +58,7 @@ if __name__ == '__main__':
     parser = argparse.ArgumentParser()
     parser.add_argument("-m", "--map_path",
                         required=True,
-                        help="Input map XML file")
+                        help="SBML map file")
     parser.add_argument("-v", "--variants_path",
                         required=True,
                         help="MINERVA variant file")

map_generator/utils/trim_long_strings.py

@@ -33,8 +33,8 @@ if __name__ == '__main__':
         datefmt='%H:%M:%S')
 
     parser = argparse.ArgumentParser()
-    parser.add_argument("-i", "--input_xml",
+    parser.add_argument("-m", "--map_path",
                         required=True,
-                        help="XML file with a map")
+                        help="SBML file with a map")
     args = parser.parse_args()
-    print(trim(args.input_xml))
\ No newline at end of file
+    print(trim(args.map_path))
\ No newline at end of file

map_generator/utils/utils.py

@@ -6,10 +6,13 @@ def register_namespaces(path) -> Dict:
     namespaces = {}
     with open(path) as f:
         content = f.read()
-        sbml_tag = re.search(r'<sbml([^>]*)>', content).group(1)
-        for ns in re.finditer(r'xmlns:*([^=]*)="([^"]*)"', sbml_tag ):
+        # sbml_tag = re.search(r'<sbml([^>]*)>', content).group(1)
+        # for ns in re.finditer(r'xmlns:*([^=]*)="([^"]*)"', sbml_tag ):
+        for ns in re.finditer(r'xmlns:*([^=]*)="([^"]*)"', content):
             k = ns.group(1)
             v = ns.group(2)
+            if k in namespaces:
+                continue
             # if not name:
             #     name = "core"
             ET.register_namespace(k, v)